From atodorov at redhat.com Thu Jan 7 13:05:50 2010 From: atodorov at redhat.com (Alexander Todorov) Date: Thu, 07 Jan 2010 15:05:50 +0200 Subject: [PATCH] handle bug excluding groups Message-ID: <4B45DC2E.8040104@redhat.com> Hello, there's a bug in handling exclusion of groups in pykickstart. Automated test case template for SNAKE which contains: ks.packages.add(['*']) ks.packages.add(['- at conflicts']) is generating incorrect ks.cfg: %packages --ignoremissing * The conflicts group is a new group in RHEL5 which lists all packages conflicting with packages from the base channel. Those are for example samba3x vs. samba, freeradius2 vs. freeradius, etc. It is expected that QE will perform test cases like: @everything - at conflicts to verify that there are no unresolved dependencies or conflicts between all base channel packages. The comments in the code says "Support syntax for removing a previously included group" but I think pykickstart doesn't know that * includes packages from @conflicts as well. Here are two different patches that fix the issue (and verified that generate expected ks.cfg). This one is proposed by James Laska: --- /usr/lib/python2.5/site-packages/pykickstart/parser.py 2010-01-07 04:16:41.000000000 -0500 +++ /tmp/parser.py 2010-01-07 04:16:29.000000000 -0500 @@ -365,4 +365,3 @@ pass - else: - newExcludedSet.add(stripped[1:]) + newExcludedSet.add(stripped[1:]) else: This one is from myself: diff --git a/pykickstart/parser.py b/pykickstart/parser.py index 2e27f03..c03c666 100644 --- a/pykickstart/parser.py +++ b/pykickstart/parser.py @@ -365,6 +365,7 @@ class Packages(KickstartObject): if stripped[1] == "@": try: self.groupList = filter(lambda g: g.name != stripped[2:], self.groupList) + newExcludedSet.add(stripped[2:]) except ValueError: pass else: -- Alexander. From clumens at redhat.com Thu Jan 7 21:34:37 2010 From: clumens at redhat.com (Chris Lumens) Date: Thu, 7 Jan 2010 16:34:37 -0500 Subject: Results of a test of pykickstart with fedora-electronics-lab lickstart In-Reply-To: <8b854d6b0912030240k3515189do490e997ffe2b42bc@mail.gmail.com> References: <8b854d6b0912030240k3515189do490e997ffe2b42bc@mail.gmail.com> Message-ID: <20100107213437.GI30178@localhost.localdomain> > 1. The out.ks file includes a directive > > %include fedora-live-base.ks > > which it should not as it already has data from the parent kickstart file. This does seem like a legitimate bug - the %include directive and the contents of the include itself should not both be written out. There are two ways to fix this, then: (1) Don't write out the %include directive, or (2) Don't write out the contents of the included file. pykickstart has no way of knowing what file a piece of data came from, which makes (2) really difficult. Therefore, it seems like (1) is the easiest way to solve this problem. Does anyone see a problem with doing this? > 2. While running the script I get the warning: > > /usr/lib/python2.6/site-packages/pykickstart/commands/partition.py:249: > UserWarning: A partition with the mountpoint / has already been > defined. > warnings.warn(_("A partition with the mountpoint %s has already been > defined.") % pd.mountpoint) > > The out.ks file contains two partition directives > part / --fstype="ext4" --size=3072 #from the parent KS > part / --size=8192 #from the child KS > > which should not be the case as the part directive in the child ks > should override the part directive of the parent ks. I don't think it's pykickstart's place to decide what to do about multiply defined partitions. I can imagine that's useful information to someone, at least. And, we do raise a warning about it so you can do something if you choose to. - Chris From darn570-linux at yahoo.com Tue Jan 12 20:44:30 2010 From: darn570-linux at yahoo.com (darn570-linux at yahoo.com) Date: Tue, 12 Jan 2010 12:44:30 -0800 (PST) Subject: Whole disk LVM via kickstart Message-ID: <432223.23594.qm@web51105.mail.re2.yahoo.com> I've been working to see if it is possible to get kickstart to perform an installation in the following disk layout. /dev/sda1 = /boot? (512mb) /dev/sdb = LVM2 pv, no partitions?(LVs /, /home, /usr/local, /opt, swap, /tmp, & /var) I'm curious if anyone has been successful in accomplishing a similar setup. I figured if I did most of my disk work in the %pre section I would be able to take advantage of some of the --useexisting options on volgroup. I even attempted creating the LVs in the %pre section and?using --useexisting?with?logvol. # Begin of %pre %pre echo 64,,83,* | sfdisk -uS /dev/sda echo | fdisk /dev/sdb < References: <432223.23594.qm@web51105.mail.re2.yahoo.com> Message-ID: <4B4CEAE7.8050304@redhat.com> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 darn570-linux at yahoo.com wrote: > I've been working to see if it is possible to get kickstart to perform an installation in the following disk layout. > > /dev/sda1 = /boot (512mb) > /dev/sdb = LVM2 pv, no partitions (LVs /, /home, /usr/local, /opt, swap, /tmp, & /var) > > I'm curious if anyone has been successful in accomplishing a similar setup. > > I figured if I did most of my disk work in the %pre section I would be able to take advantage of some of the --useexisting options on volgroup. I even attempted creating the LVs in the %pre section and using --useexisting with logvol. > > # Begin of %pre > %pre > echo 64,,83,* | sfdisk -uS /dev/sda > > echo | fdisk /dev/sdb < o > w > EOF > > # create vg00 > lvm.static pvcreate /dev/sdb > lvm.static vgcreate -s 32m vg00 /dev/sdb > > # End of %pre > > All of my %pre section code is working properly and the vg is activated. My present error is "You have not defined a root partition (/), which is required for installation of Red Hat Enterprise Linux Server to continue. > > This configuration would be aimed at a virtual environment where as growth is required I can dynamically extend sdb and then follow up with rescanning and using LVM2 commands to extend. I am trying to avoid adding additional virtual disks. > > Chad No need to do anything in %pre. You can do all of this with partition commands: bootloader --location=mbr zerombr yes clearpart --all part /boot --fstype ext3 --size=512 --ondisk=sda part pv.01 --size 1 --grow --ondisk=sdb volgroup VG pv.01 logvol / --name=LV_ROOT --fstype=ext3 --vgname=VG --size=10240 logvol /var --name=LV_VAR --fstype=ext3 --vgname=VG --size=2048 logvol swap --name=LV_SWAP --fstype=swap --vgname=VG --recommended ...etc... - -- Kyle Powell | Red Hat | Senior Consultant, RHCE -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFLTOrn7pTtanQdBU4RAjMUAJ48PAeJIWwLPeyyy3CrQ/XRcSygaQCdF6DV zy4yjLKKCNvHpKAaFWnK4xk= =f8Wi -----END PGP SIGNATURE----- From darn570-linux at yahoo.com Tue Jan 12 22:59:02 2010 From: darn570-linux at yahoo.com (darn570-linux at yahoo.com) Date: Tue, 12 Jan 2010 14:59:02 -0800 (PST) Subject: Whole disk LVM via kickstart In-Reply-To: <4B4CEAE7.8050304@redhat.com> References: <432223.23594.qm@web51105.mail.re2.yahoo.com> <4B4CEAE7.8050304@redhat.com> Message-ID: <73103.99499.qm@web51105.mail.re2.yahoo.com> Kyle, I believe that code will create a partition table on on sdb so I would end up with /dev/sdb1 (id 8e) which would be the entire disk. What I'm trying to avoid is creating a partition table all together. That way when I dynamically resize sdb I don't have to deal with managing the partition table. Chad ----- Original Message ---- From: Kyle Powell To: Discussion list about Kickstart Sent: Tue, January 12, 2010 3:34:31 PM Subject: Re: Whole disk LVM via kickstart -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 darn570-linux at yahoo.com wrote: > I've been working to see if it is possible to get kickstart to perform an installation in the following disk layout. > > /dev/sda1 = /boot? (512mb) > /dev/sdb = LVM2 pv, no partitions (LVs /, /home, /usr/local, /opt, swap, /tmp, & /var) > > I'm curious if anyone has been successful in accomplishing a similar setup. > > I figured if I did most of my disk work in the %pre section I would be able to take advantage of some of the --useexisting options on volgroup. I even attempted creating the LVs in the %pre section and using --useexisting with logvol. > > # Begin of %pre > %pre > echo 64,,83,* | sfdisk -uS /dev/sda > > echo | fdisk /dev/sdb < o > w > EOF > > # create vg00 > lvm.static pvcreate /dev/sdb > lvm.static vgcreate -s 32m vg00 /dev/sdb > > # End of %pre > > All of my %pre section code is working properly and the vg is activated. My present error is "You have not defined a root partition (/), which is required for installation of Red Hat Enterprise Linux Server to continue. > > This configuration would be aimed at a virtual environment where as growth is required I can dynamically extend sdb and then follow up with rescanning and using LVM2 commands to extend. I am trying to avoid adding additional virtual disks. > > Chad No need to do anything in %pre. You can do all of this with partition commands: bootloader --location=mbr zerombr yes clearpart --all part /boot --fstype ext3 --size=512 --ondisk=sda part pv.01 --size 1 --grow --ondisk=sdb volgroup VG pv.01 logvol /? ? --name=LV_ROOT --fstype=ext3 --vgname=VG --size=10240 logvol /var --name=LV_VAR? --fstype=ext3 --vgname=VG --size=2048 logvol swap --name=LV_SWAP --fstype=swap --vgname=VG --recommended ...etc... - -- Kyle Powell | Red Hat | Senior Consultant, RHCE -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iD8DBQFLTOrn7pTtanQdBU4RAjMUAJ48PAeJIWwLPeyyy3CrQ/XRcSygaQCdF6DV zy4yjLKKCNvHpKAaFWnK4xk= =f8Wi -----END PGP SIGNATURE----- _______________________________________________ Kickstart-list mailing list Kickstart-list at redhat.com https://www.redhat.com/mailman/listinfo/kickstart-list From muellejo at aschendorff.de Wed Jan 13 08:21:22 2010 From: muellejo at aschendorff.de (Jodok Ole Muellers) Date: Wed, 13 Jan 2010 09:21:22 +0100 Subject: Whole disk LVM via kickstart In-Reply-To: <432223.23594.qm@web51105.mail.re2.yahoo.com> References: <432223.23594.qm@web51105.mail.re2.yahoo.com> Message-ID: <20100113092122.5254c144@jodok.aschendorff.de> Hi Chad, > /dev/sdb = LVM2 pv, no partitions?(LVs /, /home, /usr/local, /opt, swap, /tmp, & /var) I recently learned that it is not a good idea not to have a partition, because you then can't define a partition type 8e and therefore e.g. booting a rescue system could get trouble to automatically identify sdb as LVM PV. This is what I use bootloader --location=mbr --driveorder=sdb --append="console=ttyS1,115200n8 console=tty1 vga=791" clearpart --all --drives=sdb part /boot --fstype ext3 --size=100 --ondisk=sdb part pv.2 --size=0 --grow --ondisk=sdb volgroup rootvg --pesize=32768 pv.2 logvol swap --fstype swap --name=swaplv --vgname=rootvg --size=8192 logvol / --fstype ext3 --name=rootlv --vgname=rootvg --size=20480 cheers From darn570-linux at yahoo.com Wed Jan 13 19:42:52 2010 From: darn570-linux at yahoo.com (darn570-linux at yahoo.com) Date: Wed, 13 Jan 2010 11:42:52 -0800 (PST) Subject: Whole disk LVM via kickstart In-Reply-To: <20100113092122.5254c144@jodok.aschendorff.de> References: <432223.23594.qm@web51105.mail.re2.yahoo.com> <20100113092122.5254c144@jodok.aschendorff.de> Message-ID: <934299.97326.qm@web51104.mail.re2.yahoo.com> I was thinking about the rescue vector this morning as well. Rescue booting with RHEL 5.4 does work and everything mounts up properly under /mnt/sysimage, you do have to deal with anaconda stopping on the disk that isn't partitioned and asking for it to be initialized twice during your rescue session. There is always a *big* concern that an admin would select yes and then blow out the LVM PV. It doesn't quite feel like a whole disk method is an accepted practice yet or either the tools have not caught up to allow this practice yet. Maybe when we can finally shove /boot into an LV the toolsets will be a little more friendly to this configuration. Chad ----- Original Message ---- From: Jodok Ole Muellers To: Discussion list about Kickstart Cc: darn570-linux at yahoo.com Sent: Wed, January 13, 2010 2:21:22 AM Subject: Re: Whole disk LVM via kickstart Hi Chad, > /dev/sdb = LVM2 pv, no partitions (LVs /, /home, /usr/local, /opt, swap, /tmp, & /var) I recently learned that it is not a good idea not to have a partition, because you then can't define a partition type 8e and therefore e.g. booting a rescue system could get trouble to automatically identify sdb as LVM PV. This is what I use bootloader --location=mbr --driveorder=sdb --append="console=ttyS1,115200n8 console=tty1 vga=791" clearpart --all --drives=sdb part /boot --fstype ext3 --size=100 --ondisk=sdb part pv.2 --size=0 --grow --ondisk=sdb volgroup rootvg --pesize=32768 pv.2 logvol swap --fstype swap --name=swaplv --vgname=rootvg --size=8192 logvol / --fstype ext3 --name=rootlv --vgname=rootvg --size=20480 cheers From an_linux at live.com Fri Jan 15 19:35:28 2010 From: an_linux at live.com (Arvind Navale) Date: Fri, 15 Jan 2010 14:35:28 -0500 Subject: Installing 32 & 64 bit package Message-ID: Hi Folks, I am kickstart a server with a custom profile where I have RPM package names listed in it instead or going with Core, Base or Everything. In the kickstart configuration I am specifying initial name of the RPM and apparently see that only one RPM package gets installed (not sure if its 32 or 64 bit). So basically I wanted to know, 1. What entries are needed in kickstart config file in order to install both 32 and 64 bit RPMs. 2. When I run 'rpm -qi' for a RPM package name how can I tell which one is the information for 32 or 64 bit. For example: rpm -qa | grep bee beecrypt-3.1.0-6 beecrypt-3.1.0-6 rpm -qi beecrypt-3.1.0-6 Name : beecrypt Relocations: (not relocatable) Version : 3.1.0 Vendor: Red Hat, Inc. Release : 6 Build Date: Sun Aug 8 15:19:50 2004 Install Date: Mon Oct 2 02:59:52 2006 Build Host: dolly.build.redhat.com Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm Size : 138837 License: LGPL Signature : DSA/SHA1, Wed Jan 5 15:03:41 2005, Key ID 219180cddb42a60e Packager : Red Hat, Inc. URL : http://sourceforge.net/projects/beecrypt Summary : An open source cryptography library. Description : Beecrypt is a general-purpose cryptography library. Name : beecrypt Relocations: (not relocatable) Version : 3.1.0 Vendor: Red Hat, Inc. Release : 6 Build Date: Sun Aug 8 23:28:17 2004 Install Date: Mon Oct 2 04:04:04 2006 Build Host: porky.build.redhat.com Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm Size : 122953 License: LGPL Signature : DSA/SHA1, Wed Jan 5 15:03:37 2005, Key ID 219180cddb42a60e Packager : Red Hat, Inc. URL : http://sourceforge.net/projects/beecrypt Summary : An open source cryptography library. Description : Beecrypt is a general-purpose cryptography library. Thank you, AN _________________________________________________________________ Hotmail: Trusted email with powerful SPAM protection. http://clk.atdmt.com/GBL/go/196390707/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From chip.shabazian at bankofamerica.com Fri Jan 15 20:14:54 2010 From: chip.shabazian at bankofamerica.com (Shabazian, Chip) Date: Fri, 15 Jan 2010 12:14:54 -0800 Subject: Installing 32 & 64 bit package In-Reply-To: References: Message-ID: Use --query-format with rpm to see architecture: Example: $ rpm -q --queryformat '%{NAME}%{VERSION}%{RELEASE}.%{ARCH}\n' beecrypt beecrypt4.1.210.1.1.x86_64 beecrypt4.1.210.1.1.i386 Use the same thing in your kickstart file %packages beecrypt.x86_64 beecrypt.i386 From: kickstart-list-bounces at redhat.com [mailto:kickstart-list-bounces at redhat.com] On Behalf Of Arvind Navale Sent: Friday, January 15, 2010 11:35 AM To: Kickstart Group Subject: Installing 32 & 64 bit package Hi Folks, I am kickstart a server with a custom profile where I have RPM package names listed in it instead or going with Core, Base or Everything. In the kickstart configuration I am specifying initial name of the RPM and apparently see that only one RPM package gets installed (not sure if its 32 or 64 bit). So basically I wanted to know, 1. What entries are needed in kickstart config file in order to install both 32 and 64 bit RPMs. 2. When I run 'rpm -qi' for a RPM package name how can I tell which one is the information for 32 or 64 bit. For example: rpm -qa | grep bee beecrypt-3.1.0-6 beecrypt-3.1.0-6 rpm -qi beecrypt-3.1.0-6 Name : beecrypt Relocations: (not relocatable) Version : 3.1.0 Vendor: Red Hat, Inc. Release : 6 Build Date: Sun Aug 8 15:19:50 2004 Install Date: Mon Oct 2 02:59:52 2006 Build Host: dolly.build.redhat.com Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm Size : 138837 License: LGPL Signature : DSA/SHA1, Wed Jan 5 15:03:41 2005, Key ID 219180cddb42a60e Packager : Red Hat, Inc. URL : http://sourceforge.net/projects/beecrypt Summary : An open source cryptography library. Description : Beecrypt is a general-purpose cryptography library. Name : beecrypt Relocations: (not relocatable) Version : 3.1.0 Vendor: Red Hat, Inc. Release : 6 Build Date: Sun Aug 8 23:28:17 2004 Install Date: Mon Oct 2 04:04:04 2006 Build Host: porky.build.redhat.com Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm Size : 122953 License: LGPL Signature : DSA/SHA1, Wed Jan 5 15:03:37 2005, Key ID 219180cddb42a60e Packager : Red Hat, Inc. URL : http://sourceforge.net/projects/beecrypt Summary : An open source cryptography library. Description : Beecrypt is a general-purpose cryptography library. Thank you, AN ________________________________ Hotmail: Trusted email with powerful SPAM protection. Sign up now. -------------- next part -------------- An HTML attachment was scrubbed... URL: From pbdlists at pinboard.com Fri Jan 15 20:08:13 2010 From: pbdlists at pinboard.com (pbdlists at pinboard.com) Date: Fri, 15 Jan 2010 21:08:13 +0100 Subject: Installing 32 & 64 bit package In-Reply-To: References: Message-ID: <20100115200813.GA6250@pinboard.com> Hi Arvind, On Fri, Jan 15, 2010 at 02:35:28PM -0500, Arvind Navale wrote: > > Hi Folks, > > I am kickstart a server with a custom profile where I have RPM package names listed in it instead or going with Core, Base or Everything. In the kickstart configuration I am specifying initial name of the RPM and apparently see that only one RPM package gets installed (not sure if its 32 or 64 bit). So basically I wanted to know, > > 1. What entries are needed in kickstart config file in order to install both 32 and 64 bit RPMs. I don't know the proper format by heart, but there is a way to specify the architechture. Something like pkgname.i386 pkgname.x86_64 Don't know, however whether the same work for package groups. if you want to install packages of a certain architecture by hand or in the %post section, you'd use: rpm -i --arch=i386 pkgname rpm -i --arch=x86_64 pkgname yum install pkganem.i386 yum install pkgname.x8664 > 2. When I run 'rpm -qi' for a RPM package name how can I tell which one is the information for 32 or 64 bit. You might want to try the following: rpm -qa --qf "%{NAME}-%{VERSION}-%{RELEASE}.%{ARCH}\n" To get a list of all the tags available use rpm --querytags Cheers, Kurt From an_linux at live.com Mon Jan 18 14:26:12 2010 From: an_linux at live.com (Arvind Navale) Date: Mon, 18 Jan 2010 09:26:12 -0500 Subject: Kickstart-list Digest, Vol 71, Issue 5 In-Reply-To: References: Message-ID: Thanks to all folks who have responded to my request! I have following question to clarify myself. I usually run 'rpm -qi pkgname' and it gives details regarding the package installed. I was wondering if there was way to tell by looking at the output of the command (rpm -qi) if its 32 or 64bit RPM package instead of using querytype option to find about it. Sorry If my question sounds incorrect. Thank you, AN From: kickstart-list-request at redhat.com Subject: Kickstart-list Digest, Vol 71, Issue 5 To: kickstart-list at redhat.com Date: Sat, 16 Jan 2010 12:00:07 -0500 Send Kickstart-list mailing list submissions to kickstart-list at redhat.com To subscribe or unsubscribe via the World Wide Web, visit https://www.redhat.com/mailman/listinfo/kickstart-list or, via email, send a message with subject or body 'help' to kickstart-list-request at redhat.com You can reach the person managing the list at kickstart-list-owner at redhat.com When replying, please edit your Subject line so it is more specific than "Re: Contents of Kickstart-list digest..." --Forwarded Message Attachment-- From: pbdlists at pinboard.com To: kickstart-list at redhat.com Date: Fri, 15 Jan 2010 21:08:13 +0100 Subject: Re: Installing 32 & 64 bit package Hi Arvind, On Fri, Jan 15, 2010 at 02:35:28PM -0500, Arvind Navale wrote: > > Hi Folks, > > I am kickstart a server with a custom profile where I have RPM package names listed in it instead or going with Core, Base or Everything. In the kickstart configuration I am specifying initial name of the RPM and apparently see that only one RPM package gets installed (not sure if its 32 or 64 bit). So basically I wanted to know, > > 1. What entries are needed in kickstart config file in order to install both 32 and 64 bit RPMs. I don't know the proper format by heart, but there is a way to specify the architechture. Something like pkgname.i386 pkgname.x86_64 Don't know, however whether the same work for package groups. if you want to install packages of a certain architecture by hand or in the %post section, you'd use: rpm -i --arch=i386 pkgname rpm -i --arch=x86_64 pkgname yum install pkganem.i386 yum install pkgname.x8664 > 2. When I run 'rpm -qi' for a RPM package name how can I tell which one is the information for 32 or 64 bit. You might want to try the following: rpm -qa --qf "%{NAME}-%{VERSION}-%{RELEASE}.%{ARCH}\n" To get a list of all the tags available use rpm --querytags Cheers, Kurt _________________________________________________________________ Your E-mail and More On-the-Go. Get Windows Live Hotmail Free. http://clk.atdmt.com/GBL/go/196390709/direct/01/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From bschneiders at woti.com Mon Jan 25 16:36:27 2010 From: bschneiders at woti.com (Bryan Schneiders) Date: Mon, 25 Jan 2010 11:36:27 -0500 Subject: Installing 32 & 64 bit package In-Reply-To: References: Message-ID: <4B5DC88B.80907@woti.com> Be aware that in Fedora 11 many packages are i586 instead of i386, and in Fedora 12 they are i686. Bryan Schneiders bschneiders at woti.com On 01/15/2010 03:14 PM, Shabazian, Chip wrote: > Use --query-format with rpm to see architecture: > > Example: > > $ rpm -q --queryformat '%{NAME}%{VERSION}%{RELEASE}.%{ARCH}\n' beecrypt > > beecrypt4.1.210.1.1.x86_64 > > beecrypt4.1.210.1.1.i386 > > Use the same thing in your kickstart file > > %packages > > beecrypt.x86_64 > > beecrypt.i386 > > *From:* kickstart-list-bounces at redhat.com > [mailto:kickstart-list-bounces at redhat.com] *On Behalf Of *Arvind Navale > *Sent:* Friday, January 15, 2010 11:35 AM > *To:* Kickstart Group > *Subject:* Installing 32 & 64 bit package > > Hi Folks, > > I am kickstart a server with a custom profile where I have RPM package > names listed in it instead or going with Core, Base or Everything. In > the kickstart configuration I am specifying initial name of the RPM and > apparently see that only one RPM package gets installed (not sure if its > 32 or 64 bit). So basically I wanted to know, > > 1. What entries are needed in kickstart config file in order to install > both 32 and 64 bit RPMs. > 2. When I run 'rpm -qi' for a RPM package name how can I tell which one > is the information for 32 or 64 bit. > > For example: > > *rpm -qa | grep bee* > > beecrypt-3.1.0-6 > > beecrypt-3.1.0-6 > > *rpm -qi beecrypt-3.1.0-6* > > Name : beecrypt Relocations: (not relocatable) > > Version : 3.1.0 Vendor: Red Hat, Inc. > > Release : 6 Build Date: Sun Aug 8 15:19:50 2004 > > Install Date: Mon Oct 2 02:59:52 2006 Build Host: dolly.build.redhat.com > > Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm > > Size : 138837 License: LGPL > > Signature : DSA/SHA1, Wed Jan 5 15:03:41 2005, Key ID 219180cddb42a60e > > Packager : Red Hat, Inc. > > URL : http://sourceforge.net/projects/beecrypt > > Summary : An open source cryptography library. > > Description : > > Beecrypt is a general-purpose cryptography library. > > Name : beecrypt Relocations: (not relocatable) > > Version : 3.1.0 Vendor: Red Hat, Inc. > > Release : 6 Build Date: Sun Aug 8 23:28:17 2004 > > Install Date: Mon Oct 2 04:04:04 2006 Build Host: porky.build.redhat.com > > Group : System Environment/Libraries Source RPM: beecrypt-3.1.0-6.src.rpm > > Size : 122953 License: LGPL > > Signature : DSA/SHA1, Wed Jan 5 15:03:37 2005, Key ID 219180cddb42a60e > > Packager : Red Hat, Inc. > > URL : http://sourceforge.net/projects/beecrypt > > Summary : An open source cryptography library. > > Description : > > Beecrypt is a general-purpose cryptography library. > > > Thank you, > AN > > > ------------------------------------------------------------------------ > > Hotmail: Trusted email with powerful SPAM protection. Sign up now. > > > > > _______________________________________________ > Kickstart-list mailing list > Kickstart-list at redhat.com > https://www.redhat.com/mailman/listinfo/kickstart-list From hardik at statementofpurpose.com Thu Jan 28 21:02:59 2010 From: hardik at statementofpurpose.com (Hardik Modi) Date: Thu, 28 Jan 2010 16:02:59 -0500 Subject: CD not found when installing from external drive Message-ID: Hello All, I create custom disks to do installs on a set of machines. Some of these machines have bezels that cover the internal disk drive tray, so my assumption is that we're always doing installs from external drives with USB connectors. My kickstart files are buried in the initrd.img, so that I can use file:/ks.cfg as the location on the boot command. Now, this generally works, but every now and then I see a machine where the user ends up with a message titled 'CD Not Found' with further text about the CentOS CD not being found in any of the CDROM drives and asking the user to insert the disk and press OK. At the bottom, there's an OK and Back pair of buttons. Prior to this, I can see the usb-storage driver being loaded, plus cdrom.ko and a few others. We used to revert to using the internal drive in these cases, but then one of my colleagues got a little frustrated and repeatedly hit the OK and Back buttons a few times and somehow the install got going again. We've now found that this works on pretty much every system that presents us with this message. I don't have a precise sequence to follow, but selecting those two buttons upwards of 10 times seems to do the trick. Alt-F3 reveals that the usual set of drivers got loaded again. I can see many messages about cdrom.ko getting unloaded and loaded again. Any pointers about how to avoid this situation? It's not a dealbreaker, but if there was a way to clean this up, I'd certainly want to. We're using stock CentOS 5.3 images with a custom kickstart file right now. We've seen this with CentOS 4.4 in the past (not certain the workaround applied). The machines are various SuperMicro brand servers. Thanks, Hardik. -------------- next part -------------- An HTML attachment was scrubbed... URL: